AI Life Progress Coach
它不是又一个待办清单，而是把长期目标拆成每天可执行、可复盘、可解释的成长系统。任务、日记、恢复状态、周/月回顾和 AI 建议落在同一条证据链里，用户能看到自己为什么在前进、哪里卡住、下一步该做什么。
People can track tasks for a few days, but long-range progress breaks down when goals, mood, journals, reviews, and next actions live in separate tools.
I built a local-first AI life operating system that turns personal data into daily and weekly feedback loops while keeping sensitive notes under user control.
The project demonstrates a broad product surface, privacy-aware architecture, and 830+ tests around a personal AI workflow that has to be both useful and trustworthy.
- 01Designed the goal-to-review workflow and the product information architecture.
- 02Implemented the Next.js application, local data model, encryption path, and AI review surfaces.
- 03Balanced motivational UX with privacy constraints so the app can be useful without exposing raw journals.
Personal productivity tools often stop at capture. This project treats long-term growth as a system of evidence: goals, daily execution, recovery, review, and AI feedback all remain connected.
Next.js app with a local-first data layer, typed domain models, and AI review modules over bounded personal context. The system separates raw private entries from shareable summaries so progress evidence can be reviewed without leaking sensitive detail.
- 01Local-first posture keeps the product credible for personal data.
- 02Daily, weekly, and monthly reviews create repeated feedback rather than one-off AI advice.
- 03Test coverage is unusually strong for a personal product prototype, which makes it useful evidence of engineering discipline.